Cancers are the leading cause of death for men and women in the USA between the ages of 40 and 79, with lung cancer causing the most mortality. However, forty-five years after the National Cancer Act of 1971 (war on cancer) was signed, progress in treating and curing cancer has been less than satisfying. Most anti-cancer therapies have debilitating toxic effects on normal tissues that eventually limit treatment. Even for the most cancer-selective therapeutics, cancers readily become drug resistant. The underlying molecular determinants that allow, drive and support cancer are quite variable, poorly understood and only recently have been used to design anti-cancer drugs. Thus, there is a significant need for improved therapeutics for cancer, including lung cancer, that exploit vulnerabilities linked to the underlying molecular basis of that particular disease. Normal cells, lacking the vulnerability, should be resistant to the cancer therapy. As an approach to identify vulnerabilities in cancer not be shared by healthy cells, we conducted a genome-wide siRNA screen of a non- small cell lung cancer (NSCLC) cell line and identified genes that were essential in this cell line, but not essential in normal cells (Tagal et al., 2017). The cell line screened, NCI-H1819, lacked expression of SMARCA4, a defect that is relatively common in several types of cancer. SMARCA4 encodes the protein Brg1, a catalytic subunit of the chromatin remodeling SWI/SNF complex, an important regulator of transcription. Through analysis of genes essential to NCI-H1819, but not normal cells, we identified a vulnerability to inhibitors of Aurora kinase A, an important mitotic kinase. The basis of the defect was an impaired centrosome- dependent mitotic spindle machinery. NSCLC cell lines that lacked Brg1 upregulate DLGAP5, which encodes the protein HURP, an essential component of a centrosome-independent machinery for mitotic spindle formation. siRNA targeting DLGAP5 was lethal for such cells, although expression of HURP is non-essential for normal somatic cells, for which the centrosome-dependent pathway of spindle formation is present. Thus, the complex containing HURP would be an ideal target for a therapeutic agent to treat cancers that have the dual biomarkers, lack of expression of Brg1 and upregulation of HURP. This application is to conduct a high- throughput screen of 300,000 drug-like chemicals to identify those that inhibit the centrosome-independent mitotic machinery, kill Brg1 null, HURP positive cancers and are non-toxic for normal somatic cells. In addition to providing potential leads for therapeutic approaches to cancer, specific chemical probes for cellular mechanisms required for cancer cell growth also provide important research tools for understanding basic cell biology or pathobiology. In the case of this application, the chemical probes generated will be important tools for understanding the molecular machinery that regulates formation of the mitotic spindle apparatus.!